Sciweavers

ICASSP
2009
IEEE
14 years 7 months ago
Combining mixture weight pruning and quantization for small-footprint speech recognition
Semi-continuous acoustic models, where the output distributions for all Hidden Markov Model states share a common codebook of Gaussian density functions, are a well-known and prov...
David Huggins-Daines, Alexander I. Rudnicky
ICASSP
2009
IEEE
14 years 7 months ago
Quickest change detection in multiple on-off processes
A Bayesian formulation of quickest change detection in multiple onoff processes is obtained within a decision-theoretic framework. For geometrically distributed busy and idle time...
Qing Zhao, Jia Ye
ICASSP
2009
IEEE
14 years 7 months ago
Data hiding in hard-copy text documents robust to print, scan and photocopy operations
This paper describes a method for hiding data inside printed text documents that is resilient to print/scan and photocopying operations. Using the principle of channel coding with...
Avinash L. Varna, Shantanu Rane, Anthony Vetro
ICASSP
2009
IEEE
14 years 7 months ago
Bispectrum on finite groups
The algebraic theory of finite groups appears in signal processing problems involving the statistical analysis of ranked data and the construction of invariants for pattern recog...
Ramakrishna Kakarala
ICASSP
2009
IEEE
14 years 7 months ago
The use of acoustically detected filled and silent pauses in spontaneous speech recognition
In recognizing spontaneous speech, the performance of typical speech recognizers tends to be degraded by filled and silent pauses, which are hesitation phenomena frequently occur...
Jun Ogata, Masataka Goto, Katunobu Itou
ICASSP
2009
IEEE
14 years 7 months ago
Covariate shift adaptation for semi-supervised speaker identification
In this paper, we propose a novel semi-supervised speaker identification method that can alleviate the influence of nonstationarity such as session dependent variation, the reco...
Makoto Yamada, Masashi Sugiyama, Tomoko Matsui
ICASSP
2009
IEEE
14 years 7 months ago
Lattice-based MLLR for speaker recognition
Maximum-Likelihod Linear Regression (MLLR) transform coefficients have shown to be useful features for text-independent speaker recognition systems. These use MLLR coefficients ...
Marc Ferras, Claude Barras, Jean-Luc Gauvain
ICASSP
2009
IEEE
14 years 7 months ago
Segmentation of malaria parasites in peripheral blood smear images
Detection of malaria parasites in stained blood smears is critical for treatment of the disease. Automation of this process will help in reducing the time taken for diagnosis and ...
Vishnu Makkapati, Raghuveer M. Rao
ICASSP
2009
IEEE
14 years 7 months ago
Time-varying Opportunistic Protocol for maximizing sensor networks lifetime
We consider transmission scheduling by medium access control (MAC) protocols for energy limited wireless sensor networks (WSN) in order to maximize the network lifetime. Time-vary...
Kobi Cohen, Amir Leshem
ICASSP
2009
IEEE
14 years 7 months ago
A Bayesian NETWORKS approach for dialog modeling: The fusion BN
Bayesian Networks, BNs, are suitable for mixed-initiative dialog modeling allowing a more flexible and natural spoken interaction. This solution can be applied to identify the in...
Fernando F. Fernández-Martínez, Javi...